[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [xri] Explicitly non-resolvable XRIs
Marc may be on to something if we use a $ tag for non-resolvability and you treat the entire xri it describes as an xref. Let's say "$$" is the non-resolvable tag. Then xri:$$.(xri:@foo.bar.baz/:1234) would say that "the xri "xri:@foo.bar.baz/:1234" is non-resolvable" (in the context of the enclosing xri). This is cool because xri:@foo.bar.baz/:1234 by itself COULD be resolvable. And there is no conflict in this (similar to Dave's suggestion of a separate scheme). This would also clarifies our equivalence issue. Because the metadata "$$" clearly describes the xri "xri:@foo.bar.baz/:1234", we distinguish equivalence at the non-metadata level from equivalence at the metadata level, which is the point I was making on the TC call yesterday (the two are apples and oranges, and you can't have equivalence rules that cross these two levels). I like this because "$$" would become the universal tag for any form of XRI that was not intended to be resolved, but had some other purpose (such as a processing instruction). It could describe an entire XRI, or just describe an xref segment within an XRI. =Drummond -----Original Message----- From: Dave McAlpin [mailto:dave.mcalpin@epokinc.com] Sent: Friday, July 11, 2003 12:04 PM To: xri@lists.oasis-open.org; Marc LeMaitre Subject: RE: [xri] Explicitly non-resolvable XRIs I don't think version (or any $ tag) would work for non-resolvability because it applies to segments rather than the XRI as a whole. It's also not available in the URI-authority portion which is a place it seems like we'd need it. I suppose you could do something like xri:(xri://foo.com).($n) to show that xri://foo.com is unresolvable, but that doesn't seem ideal to me. I may also just be misunderstanding your proposal. Dave -----Original Message----- From: Marc LeMaitre [mailto:marc.lemaitre@onename.com] Sent: Thursday, July 10, 2003 7:54 PM To: Dave McAlpin; xri@lists.oasis-open.org Subject: RE: [xri] Explicitly non-resolvable XRIs Could you put meta data in the version syntax that indicated that the XRI points to an unresolvable resource? That way an identifier could point to a non-resolvable resource in one version and a resolvable resource the next without having to change the main body of the XRI. Marc -----Original Message----- From: Dave McAlpin [mailto:dave.mcalpin@epokinc.com] Sent: Thursday, July 10, 2003 7:36 PM To: xri@lists.oasis-open.org Subject: [xri] Explicitly non-resolvable XRIs There's a desire to provide syntax that marks an XRI as explicitly unresolvable. Draft-06 includes a bang "!" character in the authority-part production for this purpose. As noted in the issues list, this causes several problems with XRIs that contain a URI-authority. For example, software that knows how to handle generic URIs will see the ! as part of the server component (most likely the hostname component), where it's an illegal character. I'm suggesting we define a different scheme, xrin, that's identical to xri except that it explicitly disallows resolution. Thoughts or comments? Dave You may leave a Technical Committee at any time by visiting http://www.oasis-open.org/apps/org/workgroup/xri/members/leave_workgroup .php
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]